Most Of Our Problem, Still, Is That We Have A Dead Battery: Felix Salmon Is Wrong Edition

|
Includes: AGG, DIA, QQQ, SPY
by: Brad DeLong

Felix Salmon is depressed:

Unemployment’s here to stay: There’s no particularly good news in these numbers. For every glimmer of good news, like the upward revisions to previous reports totaling 100,000 new jobs or so, there’s an offsetting piece of bad news, like the broad U6 unemployment rate jumping up to 16.5% from 16.2%. And the number of people unemployed for more than six months is now 6.24 million — up by 208,000. The long-term unemployed — the least employable of the unemployed, and the most intractable problem in terms of getting America back to work — are now 44.6% of the total, up from 42.9% last month, and 41.8% a year ago….

[I]nsofar as employers are hiring new people, they’re hiring new entrants into the labor force, rather than people making up the ranks of the unemployed. Maybe it’s recent graduates, maybe it’s former stay-at-home moms who were never claiming unemployment but who are now getting jobs. Maybe it’s immigrants. But the big picture is that employment growth is more or less keeping track with population growth, leaving no new jobs for the 14 million unemployed Americans.

It’s worth asking, in this context, whether Obama’s jobs bill would actually change that dynamic at all. It might help at the margin….

Is there anything the government can do to bring unemployment down? Or is it now too late? If we are indeed in the early months of a double-dip recession, than I think it is too late: unemployment is more likely to go up than it is down from here. And even if the economy’s still managing to eke out modest growth, I don’t see much hope that the unemployment rate will come down to a remotely acceptable level any time soon. Realistically, America’s unemployed are here to stay. And we’re only just beginning to understand how that’s going to affect the political economy of the nation.

The first thing to note is the JOLTS data on hiring, firing, and quits:

FRED Graph  St Louis Fed 7

In a normal month in the American economy something like 220,000 people enter the labor force, 90,000 people retire, and 5,000,000 people are hired. The amount of "churn" in the labor market is immense. Right now we still have about 220,000 people per month entering the labor force, 90,000 retiring, but only about 4,000,000 people a month are being hired. Of those 4,000,000 per month, only about 200,000 are new entrants into the labor force and only about 1,800,000 are people who quit. That still leaves 2,000,000 people per month hired from the ranks of the unemployed and of those who have just been fired.

The labor market matching process is working much less well than it works in normal times. But it is not frozen due to structural problems. There is still a huge amount of motion there.

The second thing to note is the Beveridge Curve, here plotted as the relationship between the employment-to-population ratio and the vacancy rate:

Microsoft Excel

When businesses are unwilling to hire for cyclical reasons -- when the vacancy rate is low -- we would expect the employment-to-population ratio to be low. When businesses are very willing to hire for cyclical reasons -- when the vacancy rate is high -- we would expect the employment-to-population ratio to be high as well. As the demand-driven business cycle proceeds, we expect the economy to perform clockwise loops around the Beveridge Curve, finding itself below the Beveridge Curve when the employment-to-population ratio is falling (as businesses first cut back on hiring and then the employment-to-population ratio falls), and above the Beveridge Curve when the employment-to-population ratio is rising.

The thing to watch out for as far as structural unemployment is concerned is an upward shift in the Beveridge Curve: an employment-to-population ratio that remains stubbornly where it is even though vacancies are rising.

Are we undergoing such a shift right now? The answer is: perhaps. With the vacancy rate at its current level, we would expect an employment-to-population rate of 60.3% rather than our current 58.3%. Some of this gap is because -- we hope -- we are starting one of the clockwise loops above the Beveridge Curve as the economy recovers. Even if all of the current depression in employment relative to vacancies is due to a structural shift -- due to the long-term unemployed becoming unemployable -- the maximum plausible shift in the Beveridge Curve is not that great: at most the Lesser Depression has lowered the full-employment employment-to-population ratio in the American economy from 63% to 61.5%. That is only one-third of our current jobs shortfall.

Thus our employment problem is, still, at least 2/3 cyclical and 1/3 structural. Our big problem is, as Keynes almost wrote in the 1930s, that our battery is dead. Give the economy a jump and get the electrical system functioning again, and 2/3 of our employment problem goes away. Then we can worry about the 1/3 or less of the problem that comes from malfunctioning cylinders.